1. HIV Cure Strategies Based on Targeted Killing of HIV-Infected Cells. We recently proposed that different modes of targeted cytotoxic therapy are applicable for primary acute infection versus established chronic infection. For acute infection, relatively short-term treatment with a transiently active modality such as a recombinant immunotoxin (RIT) might result in eradication of infectious HIV. For chronic infection, continuous suppression of persisting replication destruction of newly activated reservoir cells is required possible with adoptive transfer of anti-HIV engineered T cells. A. Acute HIV infection: Env-targeted recombinant immunotoxin (RIT). In with Dr. Ira Pastan, NCI, we have engineered several highly potent RITs, targeted by either CD4 (D1D2) or an scFv from a broadly neutralizing monoclonal antibody, including the newly generated VRC07. In FY2016 we initiated a collaborative NHP study with the VRC (Dr. Richard Koup, Dr. John Mascola) and MHRP (Dr. Diane Bolton) to test whether complementation of cART with an RIT can achieve a cure of rhesus macaques acutely infected with a SHIV. Preliminary results indicate that a complete cure was not achieved, though there were some experimental anomalies requiring additional analyses. B. Chronic infection: autologous T cells genetically modified with novel anti-HIV chimeric antigen receptors (CARs) to achieve a functional cure. We seek to design highly potent CARs that approach two critical properties: inescapable and non-immunogenic. We employ CD4 as a component of the targeting motif, since all HIV-1 clinically significant variants presumably must retain CD4 binding. We add a second Env-binding domain with the goal of achieving two significant improvements: enhanced potency, and blocking of the entry receptor activity of the CD4 moiety that would otherwise render transduced CD8+ T cells susceptible to infection. In FY2016, we expanded in vitro analyses of bispecific CARs containing CD4 linked to the carbohydrate recognition domain (CRD) of various human C-type lectins. CD4-MBL2 CAR displays greatly enhanced potency compared to the monospecific CD4 CAR; moreover the CRD moiety prevents the CD4 from acting as an entry receptor. We produced all-rhesus versions and observed enhanced SIV suppression by the CD4-MBL2 compared to the CD4 CAR; however, the degree of suppression was not as great as the full suppression seen in earlier studies with the human CD4-MBL2 CAR against genetically diverse HIV-1 isolates. Based on recent reports by other groups, we are testing whether a single a.a. substitution in the CD4 moiety (I39N) will provide the needed enhancement for progression to in vivo studies. C. Mechanistic studies of CAR-T cell potency. We have initiated collaborative studies with Dr. R. Brad Jones (GWU School of Medicine). The goal is to compare the CAR activities with the inefficient suppression observed with autologous CD8+ CTL clones from the same individuals. In particular, we will test the possibility that CAR-T cells can kill infected cells prior to their synthesis of new Env, by recognizing Env introduced by the incoming virion or Env on virion bound to target cells prior to entry. D. Targeting of CAR-T cells to B cell follicles. The poor accessibility of the B cell follicle to CD8+ CTL is believed to be a major contributing factor to HIV persistence in peripheral lymph nodes of cART-suppressed individuals and elite controllers. In collaboration with Dr. Pamela Skinner (U. Minnesota) and Dr. Elizabeth Connick (U. Arizona), we are expressing CXCR5, the follicle-homing chemokine receptor, on CAR-transduced T cells. Confirmation of CXCR5 expression and functional activity will support advancing to studies in rhesus macaques. 2. Structure/Function Studies of HIV-1 Env-Receptor Interactions A. Design of soluble protein constructs displaying the extracellular domains of the HIV coreceptors CCR5 and CXCR4. The HIV-1 Env glycoprotein has been studied extensively at the atomic level (X-ray crystallography and high resolution cryo-EM), including constructs of the gp120 glycoprotein (the receptor-binding subunit) in unliganded form and bound to soluble CD4 and specific monoclonal antibodies. However, the complex of gp120 bound to coreceptor (CCR5 or CXCR4) has thus far proven refractory to such analyses, in large part due to the intricate membrane associations of these molecules (they are GPCRs with 7 TM segments). In FY2016, we designed and studied a soluble chimeric protein construct (termed sCD4-CCR5nt) containing CD4 (D1D2) linked to the CCR5 N-terminal region, with optimized linker length and co-expressed with the human tyrosylprotein sulfotransferase-2 (TPST-2) to achieve the required tyrosine sulfation of the CCR5 domain. The CD4 moiety was found by ELISA to have normal binding to gp120. The sCD4-CCR5nt chimeric protein was compared to soluble CD4 is two functional assays, where the expected converse activities were observed. The first was the soluble CD4-activated fusion assay (quantitative reporter gene readout), which measures fusion between effector cells expressing Env and target cells expressing CCR5 but lacking CD4; no fusion is observed unless Env is activated to exposed the coreceptor binding site. Whereas sCD4 induced the expected strong induction of fusion, this did not occur with sCD4-CCR5nt; a control protein consisting of sCD4 linked to a variant of the CCR5 N-terminus in which the tyrosines were mutated to alanine, showed fusion induction. The simplest interpretation is that the sulfated CCR5 N-terminal moiety of sCD4-CCR5nt bound to its cognate binding site on the activated gp120, thereby preventing its functional interaction with the cellular CCR5. In a second assay, i.e. neutralization of HIV-1 pseudovirus infection, the opposite pattern was observed: sCD4-CCR5nt showed significantly higher fusion inhibition potency compared to sCD4; the chimeric protein with the Tyr to Ala mutations was less potent than the wild-type chimeric protein. Again, these data support the interpretation that the CCR5nt domain of the chimeric protein bound to its cognate site on gp120, thereby preventing its interaction with cellular CCR5 and enhancing neutralizing activity. Finally, binding competition studies (Octet) with a monoclonal antibody known to bind to the region of gp120 normally targeted by the CCR5 N-terminus confirmed the binding of the CCR5 domain of sCD4-CCR5nt with that region. These results support the potential utility of this chimeric protein in structural studies to define the gp120/CCR5 interaction at the atomic level (X-ray crystallography, in collaboration with Dr. Priyamveda Acharya). B. Molecular Mimicry in the HIV-1 Env-CD4 Interaction. In collaboration with Dr. Paolo Lusso, NIAID, we have continued studies based on awareness of a short stretch of amino acids that is common to both HIV-1 (highly conserved) and CD4. Based on published crystallographic studies of the gp120-CD4 complex, we proposed a novel gp120 model that contributes to the native structure of the unliganded HIV-1 trimer. A variety of experimental strategies are in progress to test this model.